Adaptive FIR Equalizer Using Support Samples for Low-Latency Taps
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing communication systems face challenges in efficiently removing artifacts and distortions in received signals, such as inter-symbol interference and cross-talk, which affect transmission throughput.
Innovation Solution
An adaptive equalizer system utilizing a finite impulse response (FIR) filter and a digital signal processor (DSP) circuit to process input digital samples, including support and data samples with different modulation formats, to generate equalized samples through adaptive tap weights based on residual error estimates.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If adaptive equalization is implemented to remove inter-symbol interference and cross-talk, then signal reception quality is improved, but device complexity increases due to the need for FIR filters and DSP circuits
Solution Approach 1:
The patent changes the parameter of tap weight adaptation by using support samples with lower-order modulation formats to generate tap weights that are then applied to equalize data samples with higher-order modulation formats. This parameter change allows the system to achieve effective equalization while reducing computational complexity compared to adapting tap weights for each sample type separately.
2Measurement precision
If tap weights are optimized for each modulation format separately, then equalization precision is improved, but processing time and latency increase
Solution Approach 1:
The patent performs preliminary action by using support samples to generate adaptive tap weights in advance, before the actual data samples need to be equalized. These pre-computed tap weights are then reused for equalizing subsequent data samples, reducing the processing time and latency while maintaining equalization precision.
3Productivity
If more processing resources are allocated to equalization, then throughput is improved, but power consumption increases
Solution Approach 1:
The patent changes the parameter of modulation format order by using support samples with lower-order modulation formats (which require less processing power) to generate tap weights for equalizing higher-order modulation format data samples. This approach improves throughput while reducing power consumption compared to processing all high-order modulation samples individually.
Data Source
Figure 1~2
Figure 3
Figure 4~5
AI summary
One example includes an equalizer system. The system includes a filter system configured to receive digital sample blocks associated with an input signal and to provide equalized digital sample blocks associated with the respective digital sample blocks based on adaptive tap weights. Each of the digital sample blocks includes samples and each of the equalized digital sample blocks includes equalized samples. The system also includes a sample set selector to select a subset of equalized samples from each of the equalized digital sample blocks at the output of the filter and an error estimator configured to implement an error estimation algorithm on the subset of the equalized samples to determine a residual error associated with the equalized samples. The system further includes a tap weight generator configured to generate the adaptive tap weights in response to the residual error and to provide the adaptive tap weights to the filter.